Automatic Transcription of pitched and unpitched sounds from polyphonic music

Overview

This code performs automatic music transcription of both pitched and unpitched sounds. The present demo includes a .wav file ("Take Five") as an example taken from the TRIOS dataset. It also includes sets of instrument templates computed using recordings from the RWC database.

If you use this code, please cite the following paper:
E. Benetos, S. Ewert, and T. Weyde, "Automatic transcription of pitched and unpitched sounds from polyphonic music", in Proc. IEEE International Conference on Acoustics, Speech, and Signal Processing, Florence, Italy, May 2014.

Related publications

E. Benetos, S. Ewert, and T. Weyde, Automatic Transcription of pitched and unpitched sounds from polyphonic music. 2014.: [More Details] [BIBT_EX] [URL (ext.)]; @proceedings{emmanouil2014a, author = {Emmanouil Benetos and Sebastian Ewert and Tillman Weyde}, title = {Automatic Transcription of pitched and unpitched sounds from polyphonic music}, booktitle = {IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)}, month = {May}, year = {2014} }

Members

Manager: Emmanouil Benetos
Developer: Sebastian Ewert, Tillman Weyde